Source: CORNELL UNIVERSITY submitted to NRP
THE FARM OF THE FUTURE: HARNESSING DATA-DRIVEN TECHNOLOGY FOR A SUSTAINABLE AND RESILIENT US AGRICULTURAL SYSTEM
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
ACTIVE
Funding Source
Reporting Frequency
Annual
Accession No.
1029659
Grant No.
2023-77038-38865
Cumulative Award Amt.
$4,310,184.00
Proposal No.
2023-00224
Multistate No.
(N/A)
Project Start Date
Dec 1, 2022
Project End Date
Nov 30, 2026
Grant Year
2023
Program Code
[FOTF]- Farm Of The Future
Recipient Organization
CORNELL UNIVERSITY
(N/A)
ITHACA,NY 14853
Performing Department
(N/A)
Non Technical Summary
US and global agriculture face challenges in sustainability, productivity, profitability, environmental integrity, resilience, rural prosperity, and inclusiveness. Data-driven technologies that might address these challenges are proliferating, but in a haphazard and disconnected manner. The massive streams of data and frequent reports or alerts generated by today's agricultural data systems, all running unique software and producing data in disparate formats, overwhelm farm personnel. Data increase in variety and amount faster than models can be developed to analyze them or networks can be built to carry them. Few technologies are well tested, making their selection and incorporation into the farm suite difficult. The Farm of the Future (FotF) requires research-based insights to resolve these and related problems.The Cornell Agricultural Systems Testbed and Demonstration Site (CAST) for the FotFwill harnessdata-driven technology for a sustainable and resilient US agricultural system. CAST will consistof a networked cluster of test farms with associated facilities and personnel that will leverage the resources of Cornell University and its partners to conduct data-driven research, extension, and education under the aegis of the Cornell Institute for Digital Agriculture (CIDA).CAST will advance, evaluate, and demonstrate data-driven solutions for food systems. Amultidisciplinary team of researchers, extension specialists, and educators from Cornell University (CU) and the University of Arkansas at Pine Bluff (UAPB) will undertake a comprehensive, systems-based approach to research, extension, and education, focusing on specific field-crop and animal models to generate knowledge, experiences, and opportunities with application to these agricultural sectors. CAST will leverage existing knowledge, resources, and cross-disciplinary activities through multiway collaboration withprivate and public stakeholders in food systems.CAST will promote stakeholder engagement in a commercial-farm-like setting where technologies and practices can be tested, their data collected, integrated, and analyzed, and their effects on decisions, animals, the environment, and people discerned. CAST will also be central to extension and education, on-site and virtually: farmers, students, researchers, and other stakeholders will help shape its research agenda, and knowledge produced will be fed back to all through continuous extension and education.CAST research aims will focus ondemonstratingthe value of integrating existing and emerging data-driven technologies and practices under commercial-farm-like conditions. Research will be organized infour areas (1) Innovation in Technology and Farm Practices, (2) Data Integration, (3) Data Analytics and Decision Support, and (4) Impact Assessment.At CAST scientific groundwork for innovation, demonstration, and evaluation of data-driven technology and management practices for farming will be conducted. CAST's unique ecosystem will support integration and testing of commercially available technologies and development, deployment, and testing of technologies in the research pipeline. The economic, environmental, and social outcomes of adopting the proposed technology solutions will be quantified using a combination of economic analysis, systems modeling, and behavioral research. CAST will enhance and demonstrate the value of integrating a wide range of existing and emerging technologies and practices.Extension activities will promote exchange of knowledge between CAST and stakeholders for harnessing technology to build more sustainable, resilient, and equitable farms and communities.The stakeholder network to be developed for this project--the CAST Network for Extension and Teaching (CAST-NET)--will involve farmers, manufacturers, consultants, academic experts, and others in every stage of problem identification, planning, implementation, evaluation, and feedback. CAST-NET will provide insights about cutting-edge technology goals, actively support adoption, and build informed trust that forthcoming technologies will repay the cost and effort required to adopt them. To promote adoption of innovations developed and demonstrated at CAST, we will communicate our vision, activities, and actionable outputs to CAST-NET and other stakeholders by providing access to in-person and virtual demonstrations, testing, evaluations, and new knowledge.Education efforts at CAST willprovide real-world, hands-on educational experiences to the next generation of agricultural leaders, scientists, and professionals.Benefits offered by new technologies will be sustained by the next generation of engaged, enthusiastic, and well-prepared students. CAST's cluster of working farms, where purposeful experiences range from handling actual soils, plants, and animals to coding, device testing, and hypothesis testing, will provide rich opportunities for experiential learning. Students will create, experiment, and experience in the development, delivery, and evaluation of technologies for the FotF. We will leverage existing programs and create new educational initiatives at CU and UAPB that employ the resources of CAST. These will include a minor and coursework in digital agriculture, internships at CAST, and a student hackathon. Through these efforts students will be engaged in research, outreach, and science communication.This projectwillfulfillthe FotF program's vision of a rural testbed supporting research, extension, and education in precision agriculture, smart automation, and data connectivity. Cornell university, the University of Arkansas at Pine Bluff, and their partners will establish the CAST testbed and realize its potential for advancing climate-change mitigation, environmental health, material and economic sustainability, and the well-being of rural communities.CAST will demonstrate efficient use of resources, reduced environmental impact, and the human role in agriculture. Improved efficiencywill increase productivity while reducing costs and environmental impacts. For example, precision agriculture will matchinputs and interventions to crop requirements in time, space, type, and quantity to optimize crop genetics and productivity.Animal agriculture will achieve efficiencies through data-driven approaches that individualize feeding and handling, identify sick animals promptly, guide precision fertility interventions, and the like.Reducing agriculture's environmental impacts is imperative to mitigating climate change, ocean dead zones, soil loss, water and air pollution, insect decline, zoonosis, and other harms. Data-driven technologies and practices of conservation, regeneration, and circular economy can reduce, eliminate, or even reverse undesired effects of food production. As farms get larger with fewer people working on them, but at higher wages, data-driven technologies are increasingly used to replace low-wage labor. These technologies can improve the remaining workers' quality of life by reducing the most burdensome tasks, eliminating others, and supporting efficient decision-making. We will determinesocial, socioeconomic, and farm-level financial impacts of technology adoption and integration.CAST's long-term goal is a truly sustainable FotF: a carbon-neutral or -negative, biodiversity-enhancing, inclusive, humane farm that restores rather than consumes the basis of its own existence.
Animal Health Component
40%
Research Effort Categories
Basic
30%
Applied
40%
Developmental
30%
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
1027210310010%
4022499310010%
6011699310010%
3073410209010%
8036099301010%
4047410310010%
4023499310010%
6013499310010%
4047210209010%
2057210301010%
Goals / Objectives
We propose the Cornell Agricultural Systems Testbed and Demonstration Site (CAST), a networked cluster of test farms with associated facilities and personnel that will leverage the resources of Cornell and its partners to conduct data-driven research, extension, and education under the aegis of the Cornell Institute for Digital Agriculture (CIDA).CAST will advance, evaluate, and demonstrate data-driven solutions for food systems. Amultidisciplinary team of new and well-established researchers, extension specialists, and educators from Cornell University (CU) and the University of Arkansas at Pine Bluff (UAPB) will undertake a comprehensive, systems-based approach to research, extension, and education, focusing on specific field-crop and animal models to generate knowledge, experiences, and opportunities with application to these agricultural sectors and far beyond. Under the framework of CIDA and other Cornell initiatives, we will leverage existing knowledge, resources, and cross-disciplinary activities through multiway collaboration; private and public stakeholders in food systems will be actively engaged. CAST will promote stakeholder engagement in a commercial-farm-like setting where technologies and practices can be tested, their data collected, integrated, and analyzed, and their effects on decisions, animals, the environment, and people discerned. CAST will also be central to extension and education, on-site and virtually: farmers, students, researchers, and other stakeholders will help shape its research agenda, and knowledge produced will be fed back to all through continuous extension and education. Our Aims are:AIM 1: Research. At CAST we will conduct scientific groundwork for innovation, demonstration, and evaluation of data-driven technology and management practices for farming. CAST's unique ecosystem will support integration and testing of commercially available technologies and development, deployment, and testing of technologies in the research pipeline. The economic, environmental, and social outcomes of adopting the proposed technology solutions will be quantified using a combination of economic analysis, systems modeling, and behavioral research. CAST will enhance and demonstrate the value of integrating a wide range of existing and emerging technologies and practices. Our specific objectiveisdemonstratingthe value of integrating existing and emerging data-driven technologies and practices under commercial-farm-like conditions. To accomplish this goal research will be organized under four thrust areas:(1) Innovation in Technology and Farm Practices, (2) Data Integration, (3) Data Analytics and Decision Support, and (4) Impact Assessment.AIM 2: Extension. The stakeholder network to be developed for this project--the CAST Network for Extension and Teaching (CAST-NET)--will involve farmers, manufacturers, consultants, academic experts, and others in every stage of problem identification, planning, implementation, evaluation, and feedback. CAST-NET will provide insights about cutting-edge technology goals, actively support adoption, and build informed trust that forthcoming technologies will repay the cost and effort required to adopt them. To promote adoption of innovations developed and demonstrated at CAST, we will communicate our vision, activities, and actionable outputs to CAST-NET and other stakeholders by providing access to in-person and virtual demonstrations, testing, evaluations, and new knowledge. Exchange knowledge between CAST and stakeholders will harnesstechnology to build more sustainable, resilient, and equitable farms and communities.AIM 3: Education. Benefits offered by new technologies will be sustained by the next generation of engaged, enthusiastic, and well-prepared students. CAST's cluster of working farms, where purposeful experiences range from handling actual soils, plants, and animals to coding, device testing, and hypothesis testing, will provide rich opportunities for experiential learning. Students will create, experiment, and experience in the development, delivery, and evaluation of technologies for the FotF. We will leverage existing programs and create new educational initiatives at Cornell Universityand the University of Arkansas at Pine Bluff that employ the resources of CAST. Students will be engaged in research, outreach, and science communication. Collectively, theseactivities will provide real-world, hands-on educational experiences to the next generation of agricultural leaders, scientists, and professionals.
Project Methods
CAST will include the state-of-the-art Cornell University Ruminant Center, Cornell Teaching Dairy Barn, and Musgrave Research Farm, which comprise a large (~2,550 acres), diverse land base in NY State. Through systematic integration of data, coordinated technology testing and demonstration, and exchanges of physical materials, these rural farm units will form an advanced hub for research, extension, and education that helps the FotF to fulfill its promise. CAST will focus on field crops and dairy production as models of the US ag economy, these being among the largest sectors in volume and value and offering some of the greatest challenges and opportunities for mitigating climate change. CAST's multi-site nature is a major strength: more can be learned about technologies, practices, and intelligent systems if they are applied across operations of varying size, type, and management. CAST's two crop production units (~2,550 acres available) and two dairy herds (~825 adult cows, 500 youngstock) will generate enough data to realistically model key challenges of integration and analytics.RESEARCH. CAST's four research thrusts will support innovation through cycles of development, deployment, and evaluation of technological and data-driven breakthroughs and test and demonstrate existing and emerging technologies and practices under commercial-farm-like conditions. Research on innovation in technology and farm practices will develop, deploy, test, and demonstrate innovative technologies and management practices under working farm conditions.Under research thrust 1,Innovation in Technology and Farm Practices, Specific Objectives forTechnology-Enhanced Field Crop Productionwill develop, refine, test, and demonstrate technologies and management practices in (1) Precision management of crop inputs, (2)Cover cropping and (3) Soil amendments such as rock dust and biochar.Under Specific ObjectiveSmart Automation and Data-Driven Precision Animal Managementthis project willevolvea suite of technologies in support of precision management and automation in animal systems that can enhance animal and human health, well-being, and performance while improving farm profitability and sustainability. Specifically, research will develop and demonstrate data-driven technologies for (1) Precision feeding and nutritional managementby data-driven ration formulation and automated monitoring and management, (2) Precision health management throughdata-analytic tools using integrated sensor and non-sensor data to predict automatically and in real time health outcomes of cattle, and (3) Precision reproductive monitoring and managementthroughautomated estrous detection tools, data-driven decision support methods using integrated data,automated fertility control,and point-of-care diagnostic devices.Under research thrust 2, Data Integration,the Cornell-developed Software Defined Farm, will convert raw, uninformative data from multiple sources and in diverse formats into high-quality data streams that enable efficient, accurate data analytics, and reporting. Data generated by thehardware and software infrastructure at CAST will be used to develop fully automated processes for real-time data capture and initial processing at the source, standardization of data and metadata, transfer to intermediate and centralized data streams, automated identification of technical failures, and automated discrimination of technical failures versus biological variation.Under research thrust3,Data Analytics and Decision Support,the wealth of data generated and captured by technologies and intensive management practices at the CAST will be interpreted by machine learning algorithms (MLAs) and data analytics to be maximally useful to farmers. We will create, refine, and test several algorithms for improved decision making for soils and crops and for animals at CAST. For cattle outcomes algorithms willautomatically predict health and reproductive outcomes in real time. For field crops outcomes, we will develop AI tools for site-specific management for adjusting inputs rate on a per-zone basis and understand the main drivers of productivity for each zone.Adata integration and fusion infrastructure integrating historical and real-time sensor and non-sensor data will enable the exploration of multiple interactions between the numerous sources of variation driving outcomes of interest.Under research thrust 4,Farm, Food, and Social Systems Impact Assessment, we will build on the knowledge gained to evaluate the expected economic, animal health, environmental, and social/socioeconomic outcomes of adoption of novel data-driven and farm management practices. Specific Objectives include evaluation of (1)Farm Financial Feasibility, (2) Whole Farm Animal and Environmental Health, and (3) Social and Socioeconomic Impacts.EXTENSION.By means of an actively involved stakeholder network, knowledge transfer, and education and training, CAST will promote development, marketization, acceptance, and adoption of the technologies and methods that it develops, tests, and demonstrates. Under Specific Objective 1, we will develop the CAST Network for Extension and Teaching (CAST-NET)in which stakeholders will participate in technology development, refinement, evaluation, and demonstration.Under Specific Objective 2,Knowledge Transfer,extension activities will target knowledge transfer and communication beyond CAST-NET. TheCAST vision and activities will be communicated through virtual platforms, on-site activities, and contributions to existing PRO-Dairy and Cornell Cooperative Extension programming. A website willaggregate and link to virtual content including live and recorded video streamed that will provide virtual CAST demonstrations for stakeholders. Two virtual courses for precision crop and animal management will be built on knowledge generated at CAST.Under Specific Objective 3,Demonstration and Training, industry engagement will include access to demonstrations, hands-on testing, and evaluations for stakeholders interested in technology and management practices employed at CAST. Industry organizations thatpartner with CAST will have access to arich data collection and integration platform for testing, awhole farm impact assessment framework for evaluation of financial and environmental feasibility, and a communications and outreach program for demonstration, on-site training, and visibility.EDUCATION. Educational efforts will focus on experiential learning--creating, touching, doing, interacting, evaluating, and reflecting, with CAST as a living classroom. Under Specific Objective1,a minor in digital agriculture (DA) will be developed to help undergraduates gain new perspectives, network across the university, do research pertinent to DA, and engage with CAST and CIDA.UnderSpecific Objective 2, aVirtual Course in DA for URM Studentswill disseminate knowledge from CAST and provide URM students from outside Cornell with an opportunity to learn from cutting-edge activities in real time.Cornell and the University of Arkansas at Pine Bluff (UAPB) will collaborate on developing a DA introductory course to a virtual format.UnderSpecific Objective 3, we will develop aResearch and Extension Summer Internship Program at CAST.Thisinternship program complements classroom learning with opportunities for extended, hands-on engagement in the development and evaluation of data-driven technologies and management practices at CAST.UnderSpecific Objective 4,Engage CAST in the CIDA Hackathon, CAST will belinkedto the annual CIDA Hackathon which gathers students, faculty, and community members to develop innovative tools and analyses driving DA research.

Progress 12/01/23 to 11/30/24

Outputs
Target Audience:Our target audience has been large and diverse. We reached out to, and interacted directly or indirectly, with thousands of stakeholders through presentations in scientific, producer, and consultant meetings. We also reached out to providers of crop and dairy farming technology, researchers, extension specialists, and teachers from other institutions, and senior leadership at Cornell University and other Universities. We have also targeted agriculture stakeholders not directly involved in production and the general public through our website and other informational activities. Changes/Problems:Notable issues included delays with acquisition and installation of equipment and technologies for crop and livestock research and extension activities due to issues beyond the control of the research team. Examples of reasons for delays included order processing from companies, supply chain delays, unavailability of products, and unavailability of technicians for installation and operation of equipment. What opportunities for training and professional development has the project provided?Students including MS and PhD (n = 12) and undergraduate students (n = 7), Postdoctoral fellows (n = 3), Research associates (n = 3), technicians (n = 5), and administrative personnel (n = 5) from several research groups at Cornell involved with CAST (PDs Giordano, Ketterings, Longchamps, Erickson, Weatherspoon, De Sa, Reed, You) were involved in research, extension, or educational activities for this reporting period. Professionals (n = 2) including data scientists and programmers from the Cornell Center for Advanced Computing (CAC) were also involved with CAST. This project resulted in crossdisciplinary training of students and personnel in many aspects of agricultural production, data science, engineering, computer science, plant science, agronomy, animal and veterinary science, sociology, economics, and agricultural communication. Among others, students, research personnel, and professionals have been trained in multiple aspects of row crop and dairy production systems, engineering and design of automated tools for crop and cattle management, development and optimization of assays systems for markers of reproductive and health status of cattle, integration of data-driven tools and equipment for farm management, development, testing, and implementation of software tools for data aggregation and advanced analytics, development of software code for automated assay signal reading, and use of statistical and dairy herd and crop management software. The PDs team worked directly with all personnel on their research and training. How have the results been disseminated to communities of interest?We reached out to, and directly or indirectly interacted with, thousands of diverse stakeholders through presentations in scientific, producer, and consultant meetings. The PD, Co-PDs, members of the management team, and communication specialist presented the research, outreach, and educational activities conducted at CAST to a diverse group of stakeholders and audiences in conferences, meetings, workshops, and one-on-one meetings. Examples include farmer groups in conferences and field days (e.g., Soil Health Field Day, Aurora Farm Field Day; Drone for Crop Spaying field day), presentations for extension personnel at the Cornell Cooperative Extension In-service, presentations for scientists at the 2024 Animal Science Modelling Meeting and the Sao Paulo School of Advanced Science on Precision Livestock Farming, and presentations for diverse dairy industry stakeholders at PRO-Dairy efficiency webinars. We also continue to reach out to communities of interest through different media channels, including a CAST website (https://cals.cornell.edu/cast-farm-future); Youtube videos (https://www.youtube.com/watch?v=5fBkZxSV94o); social media, including LinkedIn and X accounts; and articles in lay magazines and other publications. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period, we will continue expanding, refining, and setting equipment as well as hardware and software infrastructure at CAST. More data-driven tools and equipment for both crops and livestock research, extension, and educational activities will be installed, operated, and integrated at CAST. We will conduct livestock research focusing on development, refinement, and validation of automated and semi-automated tools and management strategies for improving dairy herd management, profitability, and sustainability. In the next year we focus in livestock research to validate and refine both wearable and non-wearable sensors for monitoring several individual cow and group level behavioral, physiological, and performance parameters of adult cows and youngstock. Similar activities are planned for crops-related research. The data integration framework and tools for reporting and advanced data analytics will be further developed and refined. We will start reporting outcomes of interest based on integrated data and will initiate deployment of algorithms for prediction of crops and dairy cattle outcomes. We will continue developing and executing partnerships with industry. We will organize events for members of CAST-NET. Through these events, we will offer training about technologies developed and integrated at CAST, informational sessions to receive stakeholder feedback for tech development and refinement, and demonstrations of technologies and data-driven management strategies at CAST for producers, farm workers, and industry partners personnel. We will continue developing an immersive experience virtual farm and educational materials about technologies and management practices at CAST in our project website. As part of CAST educational activities, we will provide data and access to students to CAST for the Cornell Digital Agriculture Annual Hackathon, we will host research summer internships for students from Cornell, UAPB, and other minority serving institutions, and we will develop and deliver an online course focusing on Digital Agriculture and activities at CAST.

Impacts
What was accomplished under these goals? CAST infrastructure capacity building: the testbed and demonstration site comprises a cluster of 3 farms operated by Cornell University (Cornell University Ruminant Center, Cornell Teaching Dairy Barn, and Musgrave Research farm). During this reporting period we continued adding and calibrating equipment and hardware and installing and setting up software infrastructure to expand research, outreach, and education capacity for crops and livestock. Work was also conducted to initiate validation, testing, and refinement of available technologies. Livestock and crops related infrastructure include data-driven tools and equipment for high-throughput phenotyping of cow behavior, physiological, and performance parameters, and high throughput phenotyping of soil, crops, air, and water. Technologies installed for livestock related activities include (1) automated behavior and physiology monitoring systems (i.e., rumination, eating, activity, body temperature, lying behavior), (2) computer vision system for automated monitoring of locomotion score and body condition at individual cow level, (3) computer vision system for monitoring feeding management and behavior of cow groups, (4) barn air monitoring sensors to detect and monitor green-house gases (GHG), volatile organic compounds (VOCs), and particulate matter, (5) walk over scales, (6) automated systems for spot measurements of methane, carbon dioxide, oxygen, and hydrogen from individual cows, and (7) centralized computer server. Examples of technologies, equipment, and data acquired for crops activities include (1) fixed-wing drones with multiple sensing capabilities, (2) soil mapping with Veris technology, (3) yield monitoring sensors for forages and grains, (4) weather stations, and (5) FTIR sensors for manure gas production monitoring. Research at CAST: Experiments and studies for development, refinement, and demonstration of data-driven technologies and management practices continued or were completed. Examples include (1) variable application rates of crop inputs (e.g., variable seed rate application across management zones; variable rate application of biochar; interseeding of cover crops in corn using an autonomous robot), (2) Single-strip spatial evaluation approach (SSEA) using unmanned aerial vehicles and satellite imagery, (3) estimation of the environmental footprint of dairy production at CAST using the COOL tool, (4) integrating microenvironment sensors for cover crops in a single mobile system (CropEnviroProbe), (5) characterization of manure lagoon methane emissions using FTIR sensing, (6) validation and demonstration of a computer vision system (CattleEye) for locomotion and body condition scoring in dairy cattle, (7) refinement and demonstration of a point-of-care diagnostic system (Repro-Phone) for pregnancy and ovarian status determination in dairy cattle, (8) development and prototype testing of a virtual fencing system for cattle, (9) development of machine learning algorithms for predicting dairy cow health, (10) testing of an automated system (Labby) for in-line milk sample collection, (11) planned and initiated studies to evaluate effects of herd and manure management practices on barn level ammonia concentrations for calibration and validation of the ammonia emissions module in the RuFas (Ruminant Farming Systems) model. CAST initiated research partnerships with other research groups at Cornell and other companies and research institutions. Partnerships were developed with scientists conducting research to estimate the environmental footprint of crop and dairy production using ecological modeling, sensing gas emissions in dairy barns and manure lagoons, use of remote sensing for estimating evapotranspiration of row crops, on-farm weather forecasting, and estimation of individual cow feed intake using wearable sensors. Data integration and analytics: continued developing hardware and software data integration infrastructure. Finalized technology infrastructure maps and data generation and exchange maps to enable data integration and analytics. Created a postgres database and associated web application that integrates automatically or semiautomatically cropping data including soil maps, satellite imagery, drone, and forage and grain yields for current and past cropping seasons. Continued refining a digitalized system for on-farm data collection of cropping activities and manure application. A similar data integration data base and web application than the one created for crops was created for collecting dairy cattle data including individual cow and herd performance, health, reproduction, behavior, and physiological data collected by multiple wearable and non-wearable sensor systems and on-farm dairy herd management software. Developed and refined MyCow$. This precision livestock technology data integration web-based tool estimates the cash flow of individual dairy cows automatically and in real time. We initiated a framework for integrating selected data from the individual animal and crop modules to create a joint animal and crop data aggregation and reporting module. Initiated connection of crop and animal data streams with the RuFas model to enable initialization of RuFas with CAST dairy farm data. CAST partnerships: One of the missions of CAST is to foster partnerships and collaborations with diverse agricultural technology developers and providers, including start-up companies and established providers of technology to the agricultural sector. CAST offers industry partners unique infrastructure and human capital to facilitate the development, refinement, testing, and demonstration of technological innovations and data-driven best management practices. This framework offers partners different levels of access to CAST infrastructure and intellectual property, resources, and outputs through a partnership and collaboration program. During this reporting we established new or solidified existing partnerships with +15 domestic and international companies including providers of technology for crops and dairy production systems. CAST communication efforts: Continued implementing our communications strategy and developing communication materials for the CAST website, social media accounts, posters for conferences and stakeholder events. Created a "Technology at CAST" section in the CAST website to feature technologies in the pipeline or currently implemented at CAST. Submitted abstracts for presentations at the 2025 US Precision Livestock Conference. Stakeholder engagement: Continued the establishment of the CAST-NET by identifying potential members of the network. Presented the CAST concept and plan for stakeholder engagement to vendors of technologies, Faculty and extension specialists from PRO-Dairy and Cornell Cooperative Extension (CCE), and other stakeholders including dairy and crop producers at several events and small group meetings. Educational activities at CAST: continued enrolling students in the minor in Digital Agriculture for undergraduate students at Cornell University. Students completed courses and participated of ancillary activities required for the minor. This DA minor provides students new opportunities to study and do research in digital agriculture, gain new perspectives on the future of agriculture, network across the university, and engage with the Cornell Institute for Digital Agriculture (CIDA) and activities at CAST. More information about the minor can be found at: https://cals.cornell.edu/education/degrees-programs/digital-agriculture-minor. The introductory course in Digital Agriculture created for the minor is being adapted for online delivery to engage students from other minority serving (e.g., UAPB) and non-minority serving institutions with Digital Agriculture and CAST. Developed a plan to establish the CAST summer internship program for underrepresented minority students from the University of Arkansas at Pine bluff (UAPB) and other MSI.

Publications

  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2025 Citation: Giordano J. O. Data-driven technologies and management practices for improving the sustainability of reproductive management. 2025. Reproduction, Fertility and Development 37, RD24148 https://doi.org/10.1071/RD24148
  • Type: Conference Papers and Presentations Status: Published Year Published: 2024 Citation: Haowen H., Waddell J., Hansen E., Foster Reed, K. Impact of exposed manure surface area on ammonia and nitrous oxide emissions in dairy systems: Insights from the Ruminant Farm Systems Models sensitivity analysis. 2024. In Proceedings of the Sao Paulo School of Advanced Science on Precision Livestock Farming. https://sites.google.com/unesp.br/plfschool
  • Type: Conference Papers and Presentations Status: Published Year Published: 2024 Citation: H. Hu, C.A. Whitcomb, K.F. Reed. Systems Engineering approach in the development of the Ruminant Farm Systems (RuFaS) model. 2024. Animal Science Modelling Meeting. West Palm Beach, Florida.
  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2024 Citation: Sunoj, S., Polson, B., Vaish, I., Marcaida III, M., Longchamps, L., van Aardt, J., & Ketterings, Q. M. (2024). Corn grain and silage yield class prediction for zone delineation using high-resolution satellite imagery. Agricultural Systems, 218, 104009.
  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2023 Citation: Longchamps, L., & Philpot, W. (2023). Full-Season Crop Phenology Monitoring Using Two-Dimensional Normalized Difference Pairs. Remote Sensing, 15(23), 5565.
  • Type: Peer Reviewed Journal Articles Status: Published Year Published: 2024 Citation: Sharma, A., Kumar, V., & Longchamps, L. (2024). Comparative performance of YOLOv8, YOLOv9, YOLOv10, YOLOv11 and Faster R-CNN models for detection of multiple weed species. Smart Agricultural Technology, 9, 100648.
  • Type: Conference Papers and Presentations Status: Under Review Year Published: 2025 Citation: Perez, M.M. and Giordano J.O. Advancing dairy health management through integrated precision livestock technologies and machine learning. 2025. Animal science proceedings. 3rd U.S. Precision Livestock Farming Conference.
  • Type: Conference Papers and Presentations Status: Under Review Year Published: 2025 Citation: Giordano J.O. and Perez, M.M. The Cornell Agricultural Systems Testbed and Demonstration site (CAST) for the Farm of the Future. 2025. Animal science proceedings. 3rd U.S. Precision Livestock Farming Conference.


Progress 12/01/22 to 11/30/23

Outputs
Target Audience:Our target audience has been large and diverse. We reached out to, and interacted directly or indirectly, with thousands of stakeholders through presentations in scientific, producer, and consultant meetings. We also reached out to providers of crop and dairy farming technology, researchers, extension specialists, and teachers from other institutions, and senior leadership at Cornell University and other Universities. We have also targeted agriculture stakeholders not directly involved in production and the general public through our website and other informational activities. Changes/Problems:Changes and problems experienced during this reporting period included Co-PDs leaving Cornell University and the University of Arkansas at Pine Bluff due to retirements or taking on new positions at other institutions. Other issues included delays with acquisition and installation of equipment and technologies for crop and livestock research and extension activities due to issues beyond the control of the research team. Examples of reasons for delays included order processing from companies, supply chain delays, unavailability of products, and unavailability of technicians for installation and operation of equipment. What opportunities for training and professional development has the project provided?Students including MS and PhD (n = 9) and undergraduate students (n = 6), Postdoctoral fellows, Research associates, and technicians (n = 9) from several research groups at Cornell involved with CAST (PDs Giordano, Ketterings, Longchamps, Erickson, Weatherspoon, Bailey) were involved in research, extension, or educational activities for this reporting period. Professionals including data scientists and programmers from the Cornell Center for Advanced Computing (CAC) were also involved with CAST. This project resulted in cross disciplinary training of students and personnel in many aspects of agricultural production, data science, engineering, computer science, plant science, agronomy, animal and veterinary science, sociology, economics, and agricultural communication. Among others, students, research personnel, and professionals have been trained in multiple aspects of row crop and dairy production systems, engineering and design of automated tools for crop and cattle management, development and optimization of assays systems for markers of reproductive and health status of cattle, integration of data-driven tools and equipment for farm management, development, testing, and implementation of software tools for data aggregation and advanced analytics, development of software code for automated assay signal reading, and use of statistical and dairy herd and crop management software. The PDs team worked directly with all personnel on their research and training. How have the results been disseminated to communities of interest?We reached out to, and directly or indirectly interacted with, thousands of diverse stakeholders through presentations in scientific, producer, and consultant meetings. The PD, Co-PDs, members of the management team, and communication specialist presented the concept of CAST, current research, extension and educational activities, and opportunities for engagement to a diverse group of stakeholders and audiences in conferences, meetings, workshops, and one-on-one meetings. Examples include the Northeast Dairy Producers Association annual meeting, the annual workshop of the Cornell Institute for Digital Agriculture, providers of agricultural technologies (e.g., Syngenta, Lely Robotics, CattleEye, Cynomys, Smaxtec, Rowbot, Agmatix), multistate research groups focused on agricultural engineering, Cornell PRO-DAIRY, dairy farming stakeholder groups, and the Cornell University College of Agriculture and Life Sciences Advisory Board. We also reached out to communities of interest through different media channels, including a CAST website (https://cals.cornell.edu/cast-farm-future); Youtube videos (https://www.youtube.com/watch?v=5fBkZxSV94o); social media, including LinkedIn and X accounts; and articles in lay magazines and other publications. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period, we will continue expanding infrastructure at CAST. More data-driven tools and equipment for both crops and livestock research, extension, and educational activities will be installed, operated, and integrated at CAST. We will conduct livestock research focusing on development, refinement, and validation of automated and semi-automated tools and management strategies for improving dairy herd management, profitability, and sustainability. Similar activities are planned for crops-related research. The data integration framework and tools for reporting and advanced data analytics will be fully functionalized. We will start reporting outcomes of interest based on integrated data and will initiate deployment of algorithms for prediction of crops and dairy cattle outcomes. We will continue developing and executing partnerships with industry. The network for extension and teaching (CAST-NET) will be fully established. As part of the network's mission, we will conduct training about technologies developed and integrated at CAST, informational sessions to receive stakeholder feedback for tech development and refinement, and demonstrations of technologies and data-driven management strategies at CAST for producers, farm workers, and industry partners personnel. We will develop an immersive experience through a virtual farm and educational materials about technologies and management practices at CAST in our project website. As part of CAST educational activities, we will provide data and access to students to CAST for the Cornell Digital Agriculture Annual Hackathon, we will host research summer internships for students from Cornell, UAPB, and other minority serving institutions, and we will develop and deliver an online course focusing on Digital Agriculture and activities at CAST.

Impacts
What was accomplished under these goals? CAST leadership and management structure: Assembled the executive board and CAST management team. Assigned roles and duties within the Executive Board and farm operations managers. Hired an Assistant CAST manager and Communication specialist. Formalized and established a partnership with collaborators at the University of Arkansas at Pine Bluff (UAPB). CAST infrastructure capacity building: the testbed and demonstration site comprises a cluster of 3 farms operated by Cornell University (Cornell University Ruminant Center, Cornell Teaching Dairy Barn, and Musgrave Research farm). Mapped existing equipment and data driven tools at each farm. Added infrastructure for conducting research, extension, and education activities for crops and livestock. Livestock and crops related infrastructure include data-driven tools and equipment for high-throughput phenotyping of cow behavior, physiological, and performance parameters, and high throughput phenotyping of soil, crops, air, and water to generate data for experimentation and reporting. Technologies installed for livestock related activities include (1) neck- and ear-attached, and ruminal boluses-based automated behavior and physiology monitoring systems (i.e., rumination, eating, activity, body temperature, water consumption, lying behavior), (2) computer vision system for automated monitoring of locomotion score and body condition scores at individual cow level, (3) computer vision system for monitoring feeding management and feeding behavior of cow groups, (4) barn air monitoring sensors to detect and monitor green-house gases (GHG), volatile organic compounds (VOCs), and particulate matter, (5) walk over scales for body weight collection, (6) automated systems for spot measurements of methane, carbon dioxide, oxygen, and hydrogen from individual cows, and (7) centralized computer server for operation of multiple animal monitoring and herd management software systems. Examples of technologies, equipment, and data acquired for crops activities include (1) fixed-wing drones with multiple sensing capabilities, (2) soil mapping with Veris technology, (3) yield monitoring sensors for forages and grains, (4) weather stations, and (5) FTIR sensors for manure gas production monitoring. Experiments and studies for development, refinement, and demonstration of data-driven technologies and management practices were designed and are being implemented at CAST. Examples include (1) variable application rates of crop inputs, (2) Single-strip spatial evaluation approach (SSEA) using unmanned aerial vehicles and satellite imagery, (3) estimation of the environmental footprint of dairy production at CAST using the COOL tool, (4) integrating microenvironment sensors for cover crops in a single mobile system (CropEnviroProbe), (5) characterization of manure lagoon methane emissions using FTIR sensing, (6) validation and demonstration of a computer vision system for locomotion and body condition scoring in dairy cattle, (7) refinement and demonstration of a point-of-care diagnostic system for pregnancy and ovarian status determination in dairy cattle, (8) development and prototype testing of a virtual fencing system for cattle, (9) development of machine learning algorithms for predicting dairy cow health, and (10) testing of an automated system for in-line milk sample collection. Data integration and analytics: established a team of computer scientists, data scientists, animal scientists, agronomists, and crop scientists that will generate data integration, analytics, and reporting methods. Developed and initiated execution of a comprehensive data integration plan. Created technology infrastructure maps and data generation and exchange maps to enable data integration and analytics. Collected crops relevant data including soil maps, satellite imagery, drone, and forage and grain yields for current and past cropping seasons. Established a digitalized system for on-farm data collection of cropping activities and manure application. Dairy cattle relevant data include individual cow and herd performance, health, reproduction, behavior, physiological data collected by multiple wearable and non-wearable sensor systems, and on-farm dairy herd management software. The data integration plan includes the development of data aggregation software tools including an animal, crop, and a joint animal and crop data integration module. CAST partnerships: One of the missions of CAST is to foster partnerships and collaborations with diverse agricultural technology developers and providers, including start-up companies and established providers of technology to the agricultural sector. To this end, we defined a partnerships framework to collaborate with industry. CAST offers industry partners unique infrastructure and human capital to facilitate the development, refinement, testing, and demonstration of technological innovations and data-driven best management practices. This framework offers partners different levels of access to CAST infrastructure and intellectual property, resources, and outputs through a partnership and collaboration program. A menu of options for engagement is available to potential partners spanning sponsored research and licensing to on-site and virtual demonstrations of partner generated data-driven tools or management practices. CAST communication efforts: Designed and implemented a communications strategy and developed communication materials including a brand, website, CAST description video, social media accounts, one-pager description of project, poster for conferences and stakeholder events, and slide deck of project overview. Stakeholder engagement: Initiated the establishment of the CAST-NET. Presented the CAST concept and plan for stakeholder engagement to vendors of technologies, Faculty and extension specialists from PRO-Dairy and Cornell Cooperative Extension (CCE), and other stakeholders including dairy and crop producers at several events and small group meetings. Educational activities at CAST: created and received approval for a minor in Digital Agriculture for undergraduate students at Cornell University. Students are currently enrolled in the minor taking courses and participating in ancillary activities. This minor provides students new opportunities to study and do research in digital agriculture, gain new perspectives on the future of agriculture, network across the university, and engage with the Cornell Institute for Digital Agriculture (CIDA) and activities at CAST. More information about the minor can be found at: https://cals.cornell.edu/education/degrees-programs/digital-agriculture-minor. An introductory course in Digital Agriculture was created and is now being adapted for online delivery to engage students from other minority serving (e.g., UAPB) and non-minority serving institutions with Digital Agriculture and CAST.

Publications

  • Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Giordano J. O. (August 29th, 2023). The Cornell Agricultural Systems Testbed and Demonstration Site for the Farm of the Future. Syngenta research and development team workshop. Ithaca, New York, United States.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Giordano J. O. (August 24th, 2023). The Cornell Agricultural Systems Testbed and Demonstration Site for the Farm of the Future. PRO-Dairy industry stakeholder meeting. Syracuse, New York, United States.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Giordano J. O. (June 21st, 2023). The Cornell Agricultural Systems Testbed and Demonstration Site for the Farm of the Future. Annual meeting of Multistate project W3009: Integrated Systems Research and Development in Automation and Sensors for Sustainability of Specialty Crops. Ithaca, New York, United States.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2023 Citation: Giordano J. O. (April 3rd, 2023). Digital Dairy Efforts at Cornell and the Cornell Agricultural Systems Testbed and Demonstration Site for the Farm of the Future. Annual meeting of Northeast Dairy Foods Research Center Meeting. Ithaca, New York, United States.